What is the CVSS score of CVE-2026-24164?

CVE-2026-24164 has a CVSS 3.1 base score of 8.8 (High). CVSS vector: CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H. EPSS exploitation probability: 0.0%.

Which versions of Nvidia are affected by CVE-2026-24164?

CVE-2026-24164 is a critical remote code execution vulnerability in NVIDIA BioNeMo Framework arising from insecure deserialization.

How to fix or mitigate CVE-2026-24164?

Within 24 hours: Identify and inventory all systems running NVIDIA BioNeMo Framework; restrict network access to BioNeMo services to trusted networks only; disable or isolate affected deployments if business operations permit. Within 7 days: Implement input validation and sandboxing controls for all serialized data processed by BioNeMo; review logs for suspicious deserialization activity or user reports of unexpected behavior; brief development and research teams on the restriction of untrusted data sources. Within 30 days: Monitor NVIDIA security advisories for patch release; prepare migration plan to patched version immediately upon availability; evaluate alternative frameworks or hardened configurations.

Nvidia CVE-2026-24164

EUVD-2026-17516 HIGH

Deserialization of Untrusted Data (CWE-502)

2026-03-31 nvidia

Deserialization RCE Denial Of Service Information Disclosure Nvidia

8.8

CVSS 3.1 · NVD

Severity by source

NVD PRIMARY

8.8 HIGH

AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H

Primary rating from NVD · only source for this CVE.

CVSS VectorNVD

CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H

Attack Vector

Network

Attack Complexity

Low

Privileges Required

None

User Interaction

Required

Scope

Unchanged

Confidentiality

High

Integrity

High

Availability

High

Lifecycle Timeline

EUVD ID Assigned

Mar 31, 2026 - 17:16 euvd

EUVD-2026-17516

Analysis Generated

Mar 31, 2026 - 17:16 vuln.today

CVE Published

Mar 31, 2026 - 16:24 nvd

HIGH 8.8

DescriptionCVE.org

NVIDIA BioNeMo contains a vulnerability where a user could cause a deserialization of untrusted data. A successful exploit of this vulnerability might lead to code execution, denial of service, information disclosure, and data tampering.

AnalysisAI

Insecure deserialization in NVIDIA BioNeMo Framework enables remote code execution when attackers can induce users to process malicious serialized data. This vulnerability (CWE-502) affects the BioNeMo Framework with network-reachable attack surface (AV:N) and low complexity (AC:L), requiring only user interaction (UI:R) but no authentication (PR:N). The CVSS 8.8 rating reflects critical impacts across confidentiality, integrity, and availability. No public exploit identified at time of analysis, though the deserialization vulnerability class is well-understood and commonly exploited. EPSS data not available for this CVE.

Technical ContextAI

NVIDIA BioNeMo Framework is a computational biology platform for AI-driven drug discovery and molecular simulation. The vulnerability stems from CWE-502 (Deserialization of Untrusted Data), a dangerous code pattern where applications deserialize data from untrusted sources without sufficient validation. In Python-based frameworks like BioNeMo, this commonly involves pickle, PyYAML, or similar serialization libraries that can instantiate arbitrary objects during deserialization. When user-controllable data flows into deserialization functions, attackers can craft malicious payloads containing executable code or object chains that trigger upon reconstruction. The CPE identifier (cpe:2.3:a:nvidia:bionemo_framework:*:*:*:*:*:*:*:*) indicates all versions of the framework are potentially affected, though specific version ranges require vendor advisory confirmation.

RemediationAI

Consult NVIDIA security advisory 5808 at https://nvidia.custhelp.com/app/answers/detail/a_id/5808 for official remediation guidance and patched BioNeMo Framework versions. Primary mitigation involves upgrading to the vendor-specified secure release that implements input validation and restricted deserialization mechanisms. Interim risk reduction measures include restricting BioNeMo Framework access to trusted users only, implementing network segmentation to limit framework exposure, validating and sandboxing all external data sources before processing, disabling automatic deserialization of user-supplied files, and monitoring for suspicious model loading or configuration parsing activities. Organizations unable to immediately patch should enforce strict data provenance controls and prohibit processing of untrusted serialized objects until updates are deployed.